CCD-world: Printed cct boards as CCD substrates?

[George Tylinski]

The following was posted to CCD-world:

Hi - Your question was specifically about circuit board durability upon 
thermal cycling to 100K... I can't answer THAT, but:

The G-10 form of epoxy/glass composite is often used in liquid nitrogen 
dewars for structural supports between the 77K tank and the ambient shell. 
The physical properties of FR-4 are very similar (FR stands for Fire 
Retardant). G-10 may be a superior material if flame retardance is not 
critical. The circuit board may warp upon cooing, so avoid overconstraining 
mounts. Delamination is not likely, and as long as the board is not under 
serious bending load, neither is through hole damage. The drawbacks of 
fiberglass are more like outgassing, cleanliness, and expansion mismatch.

There are other concerns that may be more significant:

Substituting silicone (a lousy thermal conductor) for beryllium (a terrific 
conductor) may make it difficult to cool the CCD. Also, the silicone itself 
has a high thermal expansion; a "large" bond area is likely to rupture at 
the periphery. Use a low-outgassing version like GE RTV 566 or 142. Few 
adhesives are spec'd down to 100K, so find out from the supplier how the 
properties change at those temps.

I don't know how forgiving the wire bonds would be, but consider the 
effects of spanning materials of vastly dissimilar thermal expansion 
coefficient, exacerbated by possible circuit board warpage and compliant 
adhesive joints.

Best Regards,
George Tylinski
Chair, Portland Area SolidWorks User Group
George Tylinski Mechanical Design & Analysis
voice/fax: 503-515-3338/3339

On Wednesday, May 03, 2000 11:24 PM, C.J.S.Damerell at rl.ac.uk 
[SMTP:C.J.S.Damerell at rl.ac.uk] wrote:
> The following was posted to CCD-world:
>
> I'm from the particle physics community, where our use of CCDs for 
tracking
> detectors has until now used beryllium substrates, with copper/kapton 
flex
> ccts bonded to them.
>
> For an R&D application, it would be more convenient for us to use 
standard
> multi-layer printed cct boards, to which the CCD would be attached with a
> silicone elastomer adhesive to avoid thermal stresses, and wire-bonded to
> the top PCB layer. For what we are doing, we would need a 4-layer board.
>
> Are there problems associated with using PCBs on repeated cycling to 
reduced
> temperature (down to say 100 K)? One could imagine delamination, breakage 
of
> through-hole plated vias,  etc.
>
> For all I know, the use of these boards in low temperature applications 
may
> be standard practice in the astronomy community. Any feedback would be 
most
> welcome.
>
> Chris Damerell.
> - -- CCD-world -- --
> CCD-world is fully moderated. Send posts to CCD-world at astro.ku.dk
> Standard replies will go to the list; address personal replies manually.
> For more information, please go to:  http://www.not.iac.es/CCD-world/
- -- CCD-world -- --
CCD-world is fully moderated. Send posts to CCD-world at astro.ku.dk
Standard replies will go to the list; address personal replies manually.
For more information, please go to:  http://www.not.iac.es/CCD-world/